Slack / Private Message Drop — page 215

[Robert Garry] Don't hold me to these precise numbers. Still cleaning up the dataset.*Cross-reactive antibodies to SARS-CoV-2 nucleoprotein, spike and receptor binding domain in pre-COVID-19 plasma and serum samples from Sierra Leonians.* A mystery regarding the COVID-19 pandemic has been the relatively low case numbers and fatality rates in most countries in Sub-Saharan Africa compared to many other countries worldwide (Mbow et al., 2020). Insufficient testing, under-reporting of cases/deaths, and population demographics must be considered, but other explanations remain open possibilities. Muller and co-workers (2007) showed that antibodies reactive to SARS-CoV antigens could be detected in 7 of 26 African bat species collected in South Africa and the Democratic Republic of the Congo (DRC) from 1986 to 1999 (Müller et al., 2007). The overall seroprevalence was approximately 7% (47/705) with the highest rates found in the fruit bat _Rousettus aegyptiacus_ (16%) and the insectivorous bat _Mops condylurus_ (12%). Antibodies to other human coronaviruses were not detected. The results were confirmed by western blotting, however, neutralizing activity to SARS-CoV or CoV nucleic acids were not found. This important study raises the possibility that a novel betacoronavirus(es) related to SARS-CoV and SARS-CoV-2 is present in African bats. To investigate possible exposure to betacoronavirus proteins in Africans we tested plasma and serum samples collected before the reporting of the first Chinese COVID-19 cases in late 2019 for the presence of antibodies that bind to SARS-CoV-2 nucleoprotein, spike or receptor binding domain (RBD). Samples were collected for a study of sequelae in Ebola survivors and their contacts in Kailahun and Kenema Districts in the Eastern Province of Sierra Leone between September 2016 and April 2019. Different ELISA-based assays to distinct SARS-CoV-2 proteins were used to mitigate against spurious reactivities to any one protein construct. Samples were tested in multiple laboratories. The ReSARSTM CoV-2 nucleoprotein (N) IgG ELISA developed by Zalgen Labs uses recombinant full-length SARS CoV-2 N produced in _Escherichia coli_ BL21(DE3) PLysS cells as the immobilized viral antigen in test wells. The Zalgen N ELISA shows >98% sensitive and specificity using serological panels from healthy United States blood donors. Plasmids for SARS-Cov-2 S protein were generated from synthetic codon-optimized DNA through sub-cloning into the pHCMV3 expression vector. RBD (residues 319-591) and "HexaPro" S (Hsieh et al., 2020) were transiently transfected into FreeStyle 293-F cells. Four days after transfection clarified supernatants were passed through a 0.22
m filter and then over StrepTactin resin (Qiagen). Proteins were eluted from the column with 2.5mM des-thiobiotin in binding buffer. The HexaPro Spike and RBD were further purified by size-exclusion chromatography. Spike and RBD ELISA showed >99% specificity when run on US normal blood samples. Significant numbers of positive ELISA reactivities to SARS-CoV-2 N, S and RBD were observed in serum and plasma samples from the Ebola survivor cohort. As expected there was a correlation between reactivities to trimeric S and RBD (p < .0001). Significant correlations were also observed between the reactivities to N and S (P=.003) or N and RBD (p < .0001). Using conservative cutoffs, 43% of samples (83/192) were positive for reactivity to either the SARS-CoV-2 N and/or the RBD. 33/192 were reactive to both N and RBD, 19/192 were reactive to N only and 31/192 were reactive to RBD only. Similar results were obtained in comparing the reactivities of N to S. Differences in reactivities to N and S or RBD are expected and similar to differential reactivities observed for LASV nucleoprotein versus glycoprotein (Heinrich et al., 2020). The distribution of the positive samples over the time of collection appears to be non-random. Approximate two-thirds (20/31, 65%) of samples collected in October 2016 contained antibodies that reacted to either N, RBD or both. Preliminary findings of positive reactivity is noted for samples collected in 2008 (20/94, 21%) and in samples from late 2019 (17/160,11%). These results suggest that the duration of antibody seroreactivity observed may be relatively short-lived as with certain other CoVs (Edridge et al., 2020; Long et al., 2020). However, we have not determined the duration of the immune responses in the subjects tested. It is unknown whether or not such responses offer any protection for SARS-CoV-2 infection or disease. Potentially, prior immunity to a SARS-like CoV in a portion of the Sierra Leonean population may mitigate against SARS-CoV-2 spread or induction of serious disease. Recent studies suggest that pre-existing immune responses against endemic human coronaviruses can mitigate disease manifestations from SARS-CoV-2 infection (Sagar et al., 2020). These results should be interpreted in the context of other recent studies that have demonstrated serologic cross-reactivity of SARS-CoV-2 with endemic and seasonal betacoronaviruses (Hicks et al., 2020). Antibodies reactive to SARS-CoV-2 S2 in SARS-CoV-2-uninfected individuals with recent HCoV infection were detected using a sensitive flow cytometry-based method (Ng et al.). Pre-existing memory B cells that are cross-reactive with other betacoronaviruses are activated on SARS-CoV-2 infection (Song et al., 2020). Cross-reactive monoclonal antibodies (mAbs) were isolated from the memory B cells. Several studies have reported T cell reactivity against SARS-CoV-2 in people with no known exposure to the virus, which may in part be related to prior exposure to seasonal coronaviruses [reviewed in (Doshi, 2020)].